Pages

Showing posts with label Network Sharing. Show all posts
Showing posts with label Network Sharing. Show all posts

Monday, 19 June 2017

Network Sharing is becoming more relevant with 5G

5G is becoming a case of 'damned if you do damned if you don't'. Behind the headlines of new achievements and faster speeds lies the reality that many operators are struggling to keep afloat. Indian and Nigerian operators are struggling with heavy debt and it wont be a surprise if some of the operators fold in due course.

With increasing costs and decreasing revenues, its no surprise that operators are looking at ways of keeping costs down. Some operators are postponing their 5G plans in favour of Gigabit LTE. Other die hard operators are pushing ahead with 5G but looking at ways to keep the costs down. In Japan for example, NTT DOCOMO has suggested sharing 5G base stations with its two rivals to trim costs, particularly focusing efforts in urban areas.


In this post, I am looking to summarise an old but brilliant post by Dr. Kim Larsen here. While it is a very well written and in-depth post, I have a feeling that many readers may not have the patience to go through all of it. All pictures in this post are from the original post by Dr. Kim Larsen.


Before embarking on any Network sharing mission, its worthwhile asking the 5W's (Who, Why, What, Where, When) and 2H's (How, How much).

  • Why do you want to share?
  • Who to share with? (your equal, your better or your worse).
  • What to share? (sites, passives, active, frequencies, new sites, old sites, towers, rooftops, organization, ,…).
  • Where to share? (rural, sub-urban, urban, regional, all, etc..).
  • When is a good time to start sharing? During rollout phase, steady phase or modernisation phase. See picture below. For 5G, it would make much more sense that network sharing is done from the beginning, i.e., Rollout Phase


  • How to do sharing?. This may sound like a simple question but it should take account of regulatory complexity in a country. The picture below explains this well:



  • How much will it cost and how much savings can be attained in the long term? This is in-fact a very important question because the end result after a lot of hard work and laying off many people may result in an insignificant amount of cost savings. Dr. Kim provides detailed insight on this topic that I find it difficult to summarise. Best option is to read it on his blog.


An alternative approach to network sharing is national roaming. Many European operators are dead against national roaming as this means the network loses its differentiation compared to rival operators. Having said that, its always worthwhile working out the savings and seeing if this can actually help.

National Roaming can be attractive for relative low traffic scenarios or in case were product of traffic units and national roaming unit cost remains manageable and lower than the Shared Network Cost.

The termination cost or restructuring cost, including write-off of existing telecom assets (i.e., radio nodes, passive site solutions, transmission, aggregation nodes, etc….) is likely to be a substantially financial burden to National Roaming Business Case in an area with existing telecom infrastructure. Certainly above and beyond that of a Network Sharing scenario where assets are being re-used and restructuring cost might be partially shared between the sharing partners.

Obviously, if National Roaming is established in an area that has no network coverage, restructuring and termination cost is not an issue and Network TCO will clearly be avoided, Albeit the above economical logic and P&L trade-offs on cost still applies.

If this has been useful to understand some of the basics of network sharing, I encourage you to read the original blog post as that contains many more details.

Futher Reading:



Wednesday, 28 May 2014

Case Study: RAN Sharing in Poland


The last post on Network sharing by NEC was surprisingly popular so I thought its worth doing a case study by Orange in Poland on how they successfully managed to share their network with T-Mobile. Full presentation embedded as follows:


Tuesday, 28 May 2013

NEC on 'Radio Access Network' (RAN) Sharing

Its been a while we looked at anything to do with Network Sharing. The last post with an embed from Dr. Kim Larsen presentation, has already crossed 11K+ views on slideshare. Over the last few years there has been a raft of announcements about various operators sharing their networks locally with the rivals to reduce their CAPEX as well as their OPEX. Even though I understand the reasons behind the network sharing I believe that the end consumers end up losing as they may not have a means of differentiating between the different operators on a macro cell.

Certain operators on the other hand offer differentiators like residential femtocells that can enhance indoor coverage or a tie up with WiFi hotspot providers which may provide them wi-fi access on the move. The following whitepaper from NEC is an interesting read to understanding how RAN sharing in the LTE would work.



Sunday, 15 July 2012

Fundamentals of Mobile Network Sharing

Some days back I blogged about the twitter discussion on 'Mobile Network Sharing'. Dr. Kim Larsen from Deutsche Telekom (DT) has now made a presentation and in his own words:


Given the renewed discussion of Network Sharing pros and cons I thought it made sense to wrap up several of my older presentations and update some of the information with latest knowledge. 


The myth of network sharing is clear -> huge savings and benefits often blinding the decision makers for the other side of the coin. 


I hope this presentation provided a fair picture of both sides of the Network Sharing Coin!


Friday, 13 July 2012

A twitter discussion on 'Integration problems with Mobile Network Sharing'






@dmavrakis: Are cost savings >> cultural and integration problems for mobile network sharing? - http://www.telecoms.com/46594/cultural-and-integration-problems-hamper-network-sharing-deals/

Dimitris Mavrakis, principal analyst at Informa Telecoms & Media, believes that a major challenge many operators should anticipate when embarking upon network sharing deals is the extent to which cultural and integration problems can slow down their progress and success. 
Mavrakis said that one such deal that he is aware of saw two operators spending several months holding meetings on a frequent basis in order to tackle cultural issues. And with some key figures in the industry, such as Orange Spain’s CTO Eduard Duato, calling for multiple vendors to share networks for cost-effective LTE rollout, cultural and integration problems will be vast. 
“If two operators cannot agree, what happens when there are three or four? The cultural problems and integration problems increase dramatically with the number of operators involved,” he said. 
He explained that there are multiple types of integration issues affecting operators in this scenario. 
“One vendor’s equipment may be compatible with the core network but with another one, it may require considerable effort to integrate. But there is a danger that after the equipment has been integrated the core network, it may still need considerable reconfiguration to work efficiently. This is just the tip of the iceberg; it could be that the billing system is not interoperable, or that the personnel are not trained to handle certain problems – there could be a million different problems.” 
Mavrakis noted the words of Graham Payne, managing director for the MBNL project, as evidence of the tribulations involved in setting up a shared network. Payne said that, unless the companies embarking on a sharing project were fully committed from board level on down, the results could be disastrous. 
“[Payne] said that the integration of the three networks was the biggest achievement of his career, and he’s a seasoned veteran – he’s not a newbie. So that speaks volumes about how difficult it can be,” said Mavrakis. 
He added that even by taking the path that Vodafone and O2 have in the UK, and splitting the market by geographic area, is not a sure-fire way to prevent such integration and cultural problems. 
“There’s no silver bullet; it really depends on each operator. In the case of Vodafone and O2 it may be a better solution, but if we’re talking about a shared network for LTE, from a cost perspective, the more operators involved, the better. The problem is working with the culture and the competitive nature of each operator.” 
However, Mavrakis did admit that because cost savings are the biggest driver for network sharing, the benefits that operators will see from such deals should be sufficient incentive to overcome any integration and cultural problems.

@KimKLarsen: (1) I have come to believe that for #NetworkSharing to be successful/sustainable it needs to provide more than "just" cost saving

@KimKLarsen: (2) Often Operators gets blinded of saving Potential and forget UPFRONT Cash requirements and Restructuring Cost needed!

@dmavrakis: Very interesting! Seems all discussion on network sharing is on benefits, but not on challenges or threats.

@KimKLarsen: (3) TMUK-H3G 3G sharing was about doing a lot more than they could standalone (for same or better TCO), EE Ltd formula similar,

@KimKLarsen: (4) PTC-Or PL Deploy incredible strong SHARED 2&3&4G network across! Poland (would not have been financially feasible standalone)

@dmavrakis: Agreed. Strong drive for single deployed nationwide network+sharing for LTE or any new RAN tech.

@KimKLarsen: (5) Negatives of ?#Networksharing? (a) upfront cash required, (b) helping competition (e.g., H3G-UK), (c) strategic lock-in,

@KimKLarsen: (6) (d) Complex governance (e) Complex & COSTLY disentanglement (f) loss of operational independence (g) asymmetric benefits etc

@dmavrakis: I would hate to think what disentanglement means in an active RAN sharing deal...

@KimKLarsen: (7) Doing more network for same/less TCO compared to standalone is a MUCH Easier case for ?#Networksharing? than just cost saving!
@KimKLarsen: ;-) There are no easy/clean Divorces! but Disentanglement of Active RAN Share w. Spectrum Sharing would be Mother of Messiness!
@dmavrakis: Is it even possible? I feel sorry for the person in charge of such a divorce :-)
@KimKLarsen: Though the lawyers would have a field day with Disentangling an Active RAN Share JV ;-) ....

@dmavrakis: Indeed! Patent litigation and network sharing divorces. Lawyers' dream cases!
@KimKLarsen: (8) Overlooked in ?#Networksharing?: (i) instant cell split=more capacity, (ii) improved coverage, (iii) Spectral efficiency boost
@dmavrakis: Very interesting! Seems all discussion on network sharing is on benefits, but not on challenges or threats.

@Gabeuk: Biggest issue I've heard for #Networksharing is the competitive dynamic -- you need to be more or less equal size

@KimKLarsen: Well then TMUK & Three UK must have been a mistake! ;-) ... Smaller party has much more to gain from sharing that's true!

@KimKLarsen: An important consideration as is the possible asymmetry! but I would not say it's the Biggest Issue nor a Blocking Stone!

@KimKLarsen: As the saying goes: "Money Makes Strange Bed Fellows" ;-) & cost savings do (often) take precedent over market dynamics.

@twehmeier: Interesting to see if agreement unwinds with transition to LTE, a la Sweden. Or are there commitments?

@Gabeuk: TMoUK and 3UK were the two smallest outfits at the time... so it worked for both of them

@Gabeuk: TeliaSonera wanted out of the deal, hence push to LTE. Telenor wont share in Norway, but will in Sweden

@Gabeuk: O2 and Vodafone are RAN sharing in the UK because they are roughly equal size; other combinations couldn't be agreed

@KimKLarsen: It is a highly asymmetric venture in terms of respective market shares and network size -> Size does Not matter too much!

@KimKLarsen: There are no other options in UK for meaningful network sharing! O2/VF needs higher net density to remain competitive!

@KimKLarsen: I just don't believe that Size Matters ;-) is a very important consideration for #Networksharing ...

@dmavrakis: Indeed, VF/O2 can't compete against MBNL's site density on a standalone basis!

@KimKLarsen: The only way O2 & VF could create a network in equal size to EEs in a meaningful economical way is to share their grids!

@Gabeuk: TMo, 3UK, and Orange all needed scale -- hence those combinations, and why VF and O2 sat aside

@Gabeuk: yes, the emergence of EE has changed the telco landscape in the UK

@KimKLarsen: (1) Changes in spectrum strength/position (low vs high freq) between partners might trigger an unwind.

@KimKLarsen: (2) though more likely than unwind would be downgrade from active to passive/site sharing!

@disruptivedean: Issue I see is around flexibility, eg if one MNO's customer profile has v different usage patterns

@dmavrakis: That may be in favor or active sharing. E.g. if busy hours or traffic patterns are different

@disruptivedean: Yes, but might be mix of signalling vs. "tonnage" intense, indoor/outdoor mix, specific geo's etc

@disruptivedean: ie network-sharing puts constraints on business models, customer targets, mktg propositions etc

@dmavrakis: Not sure if network sharing has effect beyond the network dept of MNO.

xoxoxoxoxoxo Added 13/07/2012 xoxoxoxoxoxo


@sadinmobile: Interesting - from experience "network" sharing should be strictly limited to a very short term or site/antenna sharing only..

@KimKLarsen: how do you get to that conclusion/experience? Particular the very short term statement? At odds with the economics!

@sadinmobile: After traffic levels pick up, networks will split anyway, until then it's nothing but tech issues and politics, s/t <5y

@KimKLarsen: 2+ yrs to get a shared network, spending abundant on integration & restructuring, w. <3 yrs to recover! Hmmm ;-)

@KimKLarsen: Though this said (1) I do agree that most of the Opex savings are in site related Opex incl. Ops/field services!

@KimKLarsen: (2) substantial Capex savings/avoidance can be achieved as well … doing so much more for the same amount of cash!

@KimKLarsen: and (3) network sharing is so much more than technology … it’s about marketing too … though politics! YEAHHH that too!

@sadinmobile: site/antenna sharing should def be done for many reasons. Also fibre backhaul sharing is a must for larger landscapes.

@KimKLarsen: particular last point is a really important one for LTE and HSPA+ where legacy backhaul falls short of air-interface!

@KimKLarsen: Few operators have the financial strength & infrastructure to standalone finance FTTS: backhaul sharing is important!

@KimKLarsen: And as backhaul FTTS is shared, so does site sharing jump out as a natural corollary.


Participants:
@dmavrakis = Dimitris Mavrakis
@KimKLarsen = Dr. Kim Larsen
@Gabeuk = Gabriel Brown
@twehmeier = Thomas Wehmeier
@disruptivedean = Dean Bubley
@sadinmobile = Sadin N

Wednesday, 5 December 2007

UTRAN Network Sharing



This is a new feature that the network operators are getting more interested in. The problem is that the Network side is more prepared than the handset side which still has some way to go.


The possibility of sharing part or all of the network by two or more separated commercial entities was not considered in the initial specification work of 3GPP. However, as e.g. a result of partnerships, the need for two or more operators to share common network infrastructure has become an economically desirable goal. Meanwhile, changes to public network operating licence conditions make such sharing possible from a regulatory point of view.


Some work has already been carried out in this area with the definition of the equivalent PLMN concept and partly with the introduction of Iu-Flex, but there is still the need to consolidate these activities under a coherent work plan.

Network sharing is in a way similar to what is done by MVNO's who use the host network to offer services. In network sharing case the other operator would not generally be a MVNO and probably the equipment would belong to both of the operators along with the cost and revenue earned. This is easily possible if the network is not overloaded (as in case of 3 UK) but if its a long time existing operator (like Vodafone) than they may not have enough spare capacity to allow someone else to share their infrastructure.

3GPP Specs for further reference:
3GPP TS 23.251 - Network Sharing; Architecture and functional description
3GPP TR 22.951 - Service aspects and requirements for network sharing